The procedure for creating a new participant record is as follows:
Log into Collect and go to the File menu -> New Participant.
Select the relevant study.
Once you have selected the study for which you wish to create a new participant, the Participant Identifier dialog will appear.
This dialog requires you to enter data about how this participant is affiliated with the study. Primarily, a participant is assigned to a particular centre (usually corresponding to a geographical region). Some studies may have only one center, in which case you do not have to make a choice. If the centre you have chosen has been further sub-divided into sites, you will need to select the site with which this participant is affiliated. Finally you may also need to specify the site consultant with whom this participant is affiliated.
Once you have specified this information, a Participant Identifier is presented to you by the system, at the bottom of the dialog. If the Participant Register is not being used you must record this number, and manually maintain the mapping between this identifier and the participant's identity (to maintain participant confidentiality and anonymity, the electronic system does not hold participant identifiable information, unless the Participant Register is used). You will need to know the Participant Identifier whenever you want to add to or modify their record.
Click OK when you have selected the centre, site and consultant and have the recorded the participant identifier.
The Consent dialog will then appear.
You can now confirm to the system that the participant has given consent by ticking the relevant consent items, which will be specific to each study. This step is important as if you don’t enter the correct consent it is possible that you won’t be able to add any data for the participant!
You must also enter one or two dates. The first date is for “Date of entry to study” and this must be completed for all studies. The second date is in response to a question that is specific to each study (and for some studies this question may not be specified at all, in which case you will only see “Date of entry to study” as above). The significance of this second date is that it is the date from which all scheduling of follow-up documents is calculated from. For instance 6 month follow ups would be expected to be completed 6 months after this date.
Some studies may use the openCDMS Participant Register, in which case you will be prompted to record the participant’s details later on in this process and so you will not need to maintain a separate log of this data. More information about the Participant Register can be found in the section called “Integration with the Participant Register” and in the Chapter 6, openCDMS Web.
Finally, for some studies you may be asked whether you wish to link the record to a record in another study. For details about record linking and data replication please see the separate openCDMS Data Replication Guide.
On completion of these steps the new participant has been successfully added, and will be opened in the Participant View ready for Documents to be completed. See the section called “Overview of the Participant View”for an overview of how the participant record is displayed.
To open an existing participant record, select 'File->Open Participant' from the menu. From the resulting dialog, first select the study to which the participant belongs. You will then be presented you with the Identifiers for all Participants from the Study that you are able to open.
The list of Participant Identifiers will contain both Participants that you have added, and also Participants added by your colleagues on the study who are members of the same Centres as yourself.
To narrow down the Identifiers that are displayed you may use the 'Filter by document status' option. By default this is set to 'All', so all available Identifiers are shown. If instead you select a specific Document Status here the list of Identifiers will be refreshed to only show those which contain Documents in the selected Status. For example, if you wish to amend Documents that have been rejected, selectiong 'Rejected' in the filter will cause the list of identifiers to only show Participant's whose Record contains rejected Documents.
Once you select a participant identifier in the dialog and then select 'OK' the record will be loaded into the 'Participant Window'. See the section called “Overview of the Participant View”for documentation on how a participant record is displayed.
Once you have a participant record loaded, you can open a document in one of two ways:
- double-click on the document of interest in the participant window.
- right-click on the document of interest, and select 'Open' from the resulting context menu.
See the section called “Introduction to the Document View” for on overview of how documents are displayed.
Documents are completed by answering the questions proposed in the data entry area. Some questions require a value to be typed in (a number or some free text) whereas others present a list of options from which the answer is selected. Dates are entered using a calendar control. Some questions may have additional help information associated with them, in which case a blue “question mark” help icon will appear next to the question text. Clicking this will display the help information.
Questions may have validation applied to prevent you from entering invalid data, for instance to restrict a numeric value to a certain range. If you enter an invalid value a red cross icon will appear next to the entry field, and hovering the mouse pointer over this will show the reason why the value is invalid.
Some documents have multiple sections, as shown in the Sections pane to the left of the window. You can move through theses sections using the “Forward” and “Back” buttons at the end of the section or using the arrow buttons located in the top-left of the window. When you try to move forward to the next section validation of the current section is performed and you will not be permitted to go to the next section if any validation errors are found. As well as invalid data missing values for mandatory questions will also cause validation errors.
You will be informed that validation errors were found, and all problematic values will be marked with the red cross icon.
If there are validation errors present you will not be permitted to move on to the next section, and hence will be unable to complete the document. If the validation error is due to you not knowing the value for a mandatory field then you may save the document in its current state as an 'incomplete' document that may be finished at a later date. Please see the section called “Saving Documents” for more information on how to save the document as either 'complete' or 'incomplete'.
Text and numeric data is just entered by typing it in to the entry field:
If for a mandatory question you do not know the answer then you should choose one of the “missing codes” that are used to signify missing data. To show the missing code menu click on the button with the downwards pointing arrow to the right of the field.
When a missing code has been selected it will be displayed in the entry field and the field itself will be disabled.
To enter an actual value in the field after a missing code has been selected open the missing code menu again and select “Enable editing”. The entry field will be cleared and is editable again.
If it anticipated that more than one line of text will be required then a larger text entry field will be displayed.
Note that a maximum of 255 characters of text may be entered in a standard text entry field, with 32768 characters permitted in the large text entry field.
There are several ways that dates may be entered. First, a date may be selected from a calendar by clicking on the button with the "-" symbol:
The date is selected by clicking on it in the calendar, which allows you to scroll back and forwards through months and years. You may also enter a date by just typing it directly into the entry field. When typing in a date it must be in “dd-mmm-yyyy” format (e.g. 17-Oct-2007) otherwise validation will fail.
In some situations you may not know the full date, but do know the month and year or year only. In this case you may enter a partial date. Click on the button with the downwards pointing arrow (as if selecting a missing code) and choose "Enter partial date".
The entry field will change to now consist of a dropdown list of months plus a field to enter the year. If you only know the year select “Unknown” from the month dropdown. The year must be entered with 4 digits.
To go back to entering a full date click on the button with the downwards pointing arrow and select “Enter full date”. Missing codes are selected in the same way as for text/numeric data.
The list of available options is presented as a set of radio buttons, one of which must be selected. Note that the missing codes appear at the end of the list of options.
Some options require additional text to be added. These options will be accompanied by a text entry field which will become enabled when the option is selected.
Some fields will be automatically calculated from other data that you enter. In the following example the Subtotal field is automatically calculated as the summation of the previous seven fields:
Slightly different from "normal" automatically calculated fields in that the calculation is not performed by Collect itself, but is done by an external service. As such you must be online in order to run the calculation. The calculation is initiated by pressing the "Calculate" button and is NOT automatically updated if changes are made to the inputs; in this case you must manually initiate the calculation again.
The following screenshot shows an example of entering data in tables.
All individual data fields work in the same way as normal, except selection from a list of options - the options are now presented in a dropdown list. To add a new row click the "New row" button – an empty row will be added to the bottom of the table. To delete a row click on the "trashcan" button to the right of that row. To apply the same missing code to all fields in a row click the "paper and pencil" button to the right of that row and select the relevant missing code from the resulting dialog.
Some fields may be disabled, in which case they will be greyed out and it is not possible to enter data in them.
Disabled fields may become enabled in response to the selection of a particular option for a preceding question.
If the option selected is then changed the fields will be disabled again, in which case any data entered when they were enabled will be cleared.
You can speed up some data entry tasks by using the keyboard, rather than the mouse, to move through a document.
The "tab" key can be used to move between the different data entry field. Repeatedly pressing "tab" causes the current selection to change. You can move back by holding "shift" and pressing "tab". For entries where a value has to be input (text, numeric) two presses of tab are required to move from one entry to the next. This is because – with the cursor in the field for entering text/numbers – one press of tab will activate the button for selecting a missing code. A second press of tab will move from this button to the next data entry field.
For multiple-choice questions pressing "space" causes the highlighted entry to be selected/deselected, depending on its status. Also, pressing space when a button is activate is equivalent to clicking the mouse on the button (e.g. to move to the next section press tab until the Forward button is active, then press space.
You can apply the same missing code (e.g. 'Not Applicable') to every entry on a section of a form using just three mouse clicks. Simply click the "missing code" button on the tool bar (see the section called “Toolbar”). Then select the required missing code and click "OK".
If Collect is exited whilst an document is being entered or edited is present then the data is automatically saved locally, and the next time Collect is started it will open up displaying the document, just as it was when Collect was shutdown.
Additionally, whilst data is being entered the document is automatically saved every minute. Therefore in the event of a computer or software error the data will not have to be re-entered
The "Import document" feature can be used to automatically transfer data from an original (source) file into a Collect document. This feature is not always available; it is enabled only for documents that have been provided with a mapping to an external data source.
In order to import from a source file into a blank Collect document, you first load the new, blank document in the same way as you would if you were to going to proceed with manual text entry. Once the document is loaded, the import button (as shown below) will appear as enabled (i.e. it will be in colour, not greyed-out and disabled):
If this icon appears in grey, it means that the current document has not been import-enabled. In this case, manual entry is your only option. Before proceeding, it is important to know the name and location of the source file from which the data will be imported. Once you are sure of the correct source file, then you are ready to proceed. By clicking on the import icon, a file selection dialog will appear (shown below). From here you can navigate to the file, which must be stored on your local computer, and must have one of the following file suffixes:
csv (comma-separated value formatted file)
xml (extended markup language formatted file)
After selecting the appropriate file, the import will occur automatically, and you should see that the document has been populated with entries. It is important that you visually inspect the data to make sure that the entries have been populated correctly; if the wrong import file were chosen, then it is possible that some or all entries in the document may get populated, although with incorrect data. Therefore, a visual inspection is always necessary.
If the import was only partially successful, then the number of errors that occurred will be reported to you in a pop-up dialog. In this case, the document will be treated as incomplete - visual inspection will reveal that certain entries have no data against them. In this case, it will be necessary for you to complete the remaining entries manually.
If you are working with a single-section document, simply fill in the data for all of the mandatory fields and then select the forward button in the document toolbar. Your document will then be saved as complete.
If you are working with a multiple-section document, you must fill in the data for all mandatory fields in the first section. Then you must navigate to the next section using the forward-arrow, and fill in its mandatory data. Repeat this process for all sections. When you have finished entering data for the final section, select the forward arrow to save the document as 'complete'.
Note that if some of the data is invalid, you will not be able to navigate to a subsequent section or to save the document. Refer to Figure 3.19, “Data entry screen - validation error on entry” to see how entries with validation errors are marked.
Note: if you are not working in 'Always-online' mode, then the document is not in the database yet. You must commit your data in order for it to be sent to the database.emphasis Offline users should see the section called “Committing Data to the Database” for information on how to commit data to the database.
If you are working on a Document and don't have all the data necessary to complete it then you may save it as 'Incomplete', which allows it to be opened again at a later date and finished off. To save a document as incomplete click the 'Save Incomplete' button in the document toolbar. See the section called “Toolbar”if you are not familiar with this button.
Note: This is only applicable for users working in 'Always-online' mode.
In exceptional and unexpected circumstances it is possible that an error may occur when Collect attempts to save a document to the database. If this occurs a warning dialog will be displayed to the user. The document will then be saved to your local hard disk (encrypted) so that the data is not lost; you may then try to save the document again at a later date. Documents saved in this way into an "error state" will be signified in the participant tree-view by a red icon. We recommend that you contact support as soon as possible informing them of the problem so that it may be investigated.
Once a document has been saved as 'Complete' it may be re-opened and edited as many times as required. Once you are happy with the data and do not anticipate further changes you should move the document to 'Pending' in order for the document to be reviewed by a Clinical Study Manager.
You do this be right-clicking on the document in the Participant View and selecting 'Submit for Review'; the Document will have its status changed to Pending and the icon for the Document in the Participant View will change signifying this. If you have multiple Documents in the same Study Stage that all need moving to Pending you can right-click on the Study Stage and select 'Submit all complete documents for review' from the pop-up menu:
Selecting this option will move all 'complete' documents in that study stage to 'pending approval'. You can also invoke this menu option by right-clicking on the Participant in the tree, which will then move all 'complete' documents for the Participant to 'Pending'.
Note: Only the Clinical Study Manager may review documents.
You must be on-line to review documents. Choose "File->Open Participant" from the menu to open the Record of the Participant whose Documents you wish to review (you can use 'Filter by document status' to show only Participants with Documents in the Pending status, as described in the section called “Opening Existing Participants”). The Participant's Record will be opened, and to review a Pending Document simply double-click on it in the Participant View to open it.
If you see something that you think may be wrong, you can mark it as invalid . To do this, right-click on the entry field and select "Mark entry as invalid".
A dialog will be displayed where you must enter an annotation; this is a description of why the value is invalid to aid the CRO in correcting it. Click OK to apply the annotation and mark the entry as invalid.
A yellow triangle icon will now appear adjacent to the entry field to notify that it has been marked as invalid; hovering the mouse pointer over this icon will show the annotation. If subsequently you change your mind and decide that the value is valid after all, right click on it and select "Mark entry as valid".
Once you have finished reviewing the document click the Forward button from the last section. If you have marked any entries as invalid then the document will be Rejected. The CRO(s) responsible for the data entry will be emailed notifying them of this.
If no entries have been marked as invalid then the document will be Approved.
You must be on-line to edit rejected documents. Choose "File->Open Participant" from the menu to open the Record of the Participant whose Documents you wish to correct (you can use 'Filter by document status' to show only Participants with Documents in the Rejected status, as described in the section called “Opening Existing Participants”). The Participant's Record will be opened, and to correct a Rejected Document simply double-click on it in the Participant View to open it.
When the document is loaded you can view the reasons why the document has been rejected. Move the mouse over the yellow warning triangles, and the reason why the value has been rejected will be displayed.
To amend the value, double click on the entry field with the left mouse button. This brings up the Edit Entry dialog, where you may edit the value and enter a reason for the change. Note that even if you do not change the value (i.e. you believe the value that was rejected to be correct) you must still enter a reason, which in this case will inform your Clinical Study Manager of the reason for your entry remaining the same.
When you have amended all of the rejected values click on "Forward" at the end of the document. The changes will be committed to the database and the document moved back to the Pending Approval state.
This action only applies to the Clinical Study Manager.
Proceed as in the section called “Reviewing Documents” . You can view the response or edits to a rejection by looking for the blue "i" icon. Hover the mouse over this icon to see the response from the CRO to the rejection of the value.
To confirm that the value is now OK right-click on the entry field and select "Mark as valid". If the value is still invalid right-click and select "Mark entry as invalid" - an annotation is required.
To view the complete change history of an entry right-click on it and select "View Provenance". This is useful to compare the current value with the previous (rejected) value.
If you are operating in 'Always-online' mode, you do not have to worry about committing data to the database because this happens automatically whenever you complete a document, or when you save it as 'incomplete'. However, if you are NOT operating in 'Always-online' mode, the information in this section is critical to you.
After a document has been completed and saved it is stored locally on your computer – IT IS NOT AUTOMATICALLY COMMITTED TO THE CENTRAL DATA DATABASE!
You should try to commit documents as soon as possible for the following reasons:
If the computer on which they are stored fails they might be lost.
If the computer on which they are stored is stolen they will be lost.
The recruitment statistics are not updated until a document is committed for a record.
Although data stored locally is encrypted, it is not as secure as data held in the database.
Others cannot see data stored locally and so it only has value when entered into the database.
To commit data to the central database select ‘Commit’ from the Database menu. A dialog will be launched showing you the records that may be committed. By default records with complete documents are pre-selected to be committed; you must select records with incomplete documents that you want to commit (note that if a record has both complete and incomplete documents it will appear in the list twice).
Once you have selected the records to commit click OK. Upon completion the records selected will have been saved to the central database and removed from your local computer.
From the participant window it is possible to view and change record-level participant data. In order to invoke the Participant Properties dialog, right-click on the participant record and then select 'View/Edit Participant Properties' in the resulting dialog, as shown below:
This will invoke the participant properties dialog.
From this dialog, it is possible to change the following record properties:
'Current Record Status' - this status is changed automatically as you complete documents for the participant. Once all documents in a study stage have been completed, the status change automatically to reflect this (e.g. once all baseline documents have been completed, the status will move automatically to 'Baseline Completed'. You may need to edit this status, however if a participant withdraws from a study, or if you wish to invalidate the current record (i.e. remove it from the study).
'Consent for the record held' - the entering of consent information is mandatory upon creation of a new participant. If any of these consent details change you may amend them by editing this property.
'Date of entry into study' - This date can be changed here if necessary. Also, some studies also record a second date - and this can also be amended.
'Notes' - general notes regarding the participant record can be modified.
This section only applies to studies where the Participant Register is used to store identifiable participant details.
For each study deployed on the openCDMS system it can be defined whether or not the study makes uses of the openCDMS Participant Register. The Participant Register allows identifiable participant data (e.g. name. address, etc.) to be stored electronically, linked to the study/screening code generated by openCDMS Collect.
If the Participant Register is being used, then when creating a new record for a participant (see the section called “Creating a new record for a participant”) an additional dialog will be displayed after the Consent dialog. The Participant Register dialog allows you to enter the participant’s details and on clicking the OK button these details will be uploaded to the Participant Register.
In order to upload the participant’s details to the Participant Register you must be online; if this is not the case then the participant’s details will be stored on your computer until the next time you use the Database – Commit action when they will be uploaded to the Participant Register (provided you are now online. To inform you that the participant’s details could not immediately be uploaded to the Participant Register the following dialog will be shown.
A participant is opened using the File -> Open Participant menu option (see the section called “Creating a new record for a participant”). This will allow you to select the participant identifier or use identifiable data (such as name or date of birth) to search for the participant.
The 'More..' button shown in the Open Participant dialog allows the search criteria to be widened to include all details stored in the participant register.
Participant details may also be viewed and edited using the openCDMS web interface; see Chapter 6, openCDMS Webfor further details.
You may edit the details of a Participant in the Register by opening the Participant as normal and right-clicking on their identifier in the participant view. Select the "View/Edit Participant Register" option and the Participant Register dialog will be displayed, in the same way as it is before randomization. Clicking Edit will allow the data ito be amended, and then clicking OK will save the data to the register.
Participant details may also be viewed and edited using the openCDMS web interface; see Chapter 6, openCDMS Webfor further details.
If the study is a clinical trial then the Participant Register will also be used to provide randomization. One of the documents will have been defined as being the trigger for randomization, in that after its completion you must decide whether the participant is eligible to participate in the trial or not. When this document is completed (i.e. the end of the document is reached and you choose to save it) a dialog will be displayed allowing you to choose whether the participant should be randomized or not.
If the participant is eligible for the trial click Yes – you will first be displayed the participant's details from the Participant Register for review; if all the details are correct click OK to proceed with randomization and allocate the participant to one of the treatment arms. If the details in the Participant Register are not correct click Edit; the fields in the dialog will be opened for editing. On clicking OK the amended details will be uploaded to the Participant Register before the randomization process is started. Clicking Cancel at any time aborts the randomization attempt (and discards changes to the Participant's details, if Edit was clicked) and you will be returned to the document view.
If the participant is not eligible for the trial click No – the participant will not be allocated to a treatment arm which implies that their participation in the study is over.
If you are unsure whether to click Yes or No then clicking the Cancel button will return you to the document view.
If there is a problem during the randomization process the system will not allow you to save the document. There are two common situations which will cause randomization to fail:
The application is offline, and so it cannot communicate with the Participant Register to perform randomization. In this case you should select “Go back and continue editing” from the Save Document dialog, then use Local – Save Incomplete Document to save the document to your computer. Then, when you are online open the incomplete document and try to save it again.
The participant’s details are not found in the Participant Register. This can happen if the record was created and the document that triggers randomization completed whilst you were offline. If you then open the incomplete document when online, save it and try to perform randomization again before you have done a Commit then randomization will fail, because the participant’s details are only uploaded to the Participant Register during the Commit process. So, perform a Commit then try to save the document again.
Data Replication functionality has been added to the openCDMS Collect data collection application in order to allow data for studies that share participants to be entered just once. A pair of data collection studies hosted on the openCDMS platform may be linked, with one being designated the primary study, and the other the secondary study. The documents (assessment forms) that are common between the two studies are also connected. When a new participant is added to the primary study the user is given the option to link this to an equivalent record in the secondary study. Then when data is entered into the primary record for any of the common documents it will automatically be replicated into the equivalent document in the secondary record.
To link a record in the primary dataset to a record in the secondary dataset at the point when the primary record is created the following steps should be taken.
Create the primary record as normal.
After the consent for the record has been entered an additional dialog box will be displayed asking if you wish to link the record to one in the secondary dataset:
If you do not want to link the record to one in the secondary dataset (and hence enter the participant into the secondary study too), click No and you will be taken straight to the view to enter data as normal. If you do wish to link it click Yes.
Assuming Yes was clicked, you now have to initialize the secondary record. This is essentially the same as creating a new record ordinarily, but as it is only possible to link the primary record to a new secondary record the identifier dialog does not include the options for saying whether a new or existing record is being used. Therefore you go straight into selecting the centre for the secondary record. Note that the centres shown are dependent upon the centre of the primary record; it is defined in the dataset of the primary record which centres in the secondary dataset each centre in the primary may be linked to.
To complete the process of linking the two records just complete the consent for the secondary record as usual. On successful entry of the consent, the records are linked.
Note that if at any point during the process of initializing the secondary record Cancel is clicked the primary record is still created, but it won't be linked to a secondary record.
It is also possible to link records for data replication when either the primary record, or both primary and secondary records, already exist. Note that the application must be online (able to connect to the central openCDMS data database) for the duration of this procedure.
Choose the Link Records option from the Advanced menu
A list of records belonging to primary datasets, that are not already linked to records in secondary datasets, will be displayed. Only records committed to the database will be displayed.
After selecting the primary record the identifier dialog is displayed to permit the selection of the secondary record. The identifier dialog is the same as the standard one used when adding a new document, and it is possible to select an existing record or create a new record.
Once the secondary record has been selected or created it is linked with the primary record. Data for documents in the primary record is then replicated from the primary record to the secondary record. We consider the data replication separately for both a new and an existing secondary record below.
When a new secondary record is linked to an existing primary record the data replication process is relatively straightforward. Data from the primary record for all complete documents – either stored locally or in the database – that are common between the primary and secondary datasets will be replicated to the secondary record. The replicated documents will initially be stored locally and must be committed to the database as normal. It is recommended that this is done as soon as possible.
When an existing secondary record is linked to an existing primary record the data replication process is essentially the same as for a new secondary record – except when there are conflicts. A conflict occurs when a document that is common between the primary and secondary datasets already exists in the secondary record. When the secondary record is linked to a primary record all data is entered via the primary record, therefore to ensure that the data is consistent between the two records any conflicts must be dealt with when the records are linked.
How conflicts are dealt with depends upon the exact situation, as follows.
Table 3.1. Data replication conflicts
| Document in primary record | Document in secondary record | Conflict resolution |
|---|---|---|
| Complete | Complete | The complete document in the secondary record is removed and replaced with a document replicated from primary record. User warned of this and may cancel the linking process. See Figure 3.57, “Complete document in both primary and secondary dialog” |
| No document | Complete | Two options are presented: 1.Replicate the document “backwards” from the secondary record to the primary record. 2.Delete the document from the secondary record. User may also cancel the linking process. See Figure 3.58, “Complete documents in the secondary record that aren't in the primary dialog” |
| Incomplete | Complete | Two options are presented: 1.Remove the incomplete document from the primary record and replace it with a document replicated backwards” from the secondary record to the primary record. 2.Remove the complete document from the secondary record. User may also cancel the linking process. See Figure 3.59, “Complete documents in the secondary, incomplete in the primary dialog” |
| Incomplete | Incomplete | The incomplete document in the secondary record is deleted. |
| No document | Incomplete | The data from the incomplete document in the secondary record is replicated “backwards” to the primary record; the incomplete document in the secondary record is then deleted. User warned of this and may cancel the linking process. See Figure 3.60, “Incomplete documents in the secondary will be moved to the primary dialog” |
Note that when one of the dialogs is displayed that permits a choice as to how the records will be linked the choice made here will apply to all documents at conflict in that way.
After the linking process has been completed, and whatever action necessary to resolve conflicts has been taken, it is possible (depending upon what conflicts there were) that both primary and secondary records, stored both locally and remotely, may have been altered. Therefore - if not using 'always-online' mode, in which case all data will be saved to the database automatically - it is recommended that both the primary and secondary records held locally are committed to the database immediately after they are linked.
After two records have been linked all further operations to them must be performed via the primary record, and these operations will then be mirrored to the secondary record. For example:
It is not possible to add a new document to the secondary record where that document is common between the primary and secondary datasets. Instead the following warning will be displayed.
It will not be possible to edit a complete document stored locally int he secondary record when it is a common document with the primary. The complete primary document should be edited instead and changes made to it will be replicated to the secondary.
All "review and approve" functionality i.e. Load Pending Document, Load Rejected Document, Load Approved Document may only be performed on shared documents in the primary record. Again, changes in data and also document status will be replicated to the secondary.
Note that data is not replicated to the secondary record when a document in the primary record is saved as incomplete. The data will be replicated when the document is first saved as complete, and subsequently after that.
Also, remember that any documents in the secondary record that do not have an equivalent in the primary record will need to be completed for the secondary record just the same as in a record for the secondary dataset that is not linked to one in the primary dataset. As the majority of the data for linked records is entered via the primary it is easy to forget this!
Normally once two records have been linked there is no need for any other explicit action; data is entered as normal for the 'primary' record and this gets automatically replicated to the 'secondary' record.
However, there are special circumstances in which the linked records can get out of sync i.e. the data stored in primary and secondary records is not identical. Some examples of how this might happen:
The secondary record does not have sufficient consent for documents to be replicated to it from the primary record. Consent is then added to the secondary record that will allow the documents to be replicated.
A document that has been in the primary dataset from the start of the study is added to the secondary dataset part-way through the study. For records created upto this point the data would not have been replicated to the secondary.
To make sure that two linked records are in sync use the 'Synchronize Linked Record" option from the Advanced menu. Select the primary record from the list and the system will then synchronize the data held in the primary and secondary records. Behind the scenes the process is essentially the same as what is done when existing records are linked (see the section called “Linking existing records”). After the synchronization has completed it is recommended that a Commit is done (if not in 'Always-online' mode), although this may not actually be necessary).
During the data entry process the document may be printed at any time by clicking the Print icon in the toolbar below the menus. This works in the same way as the Database – Print Document menu item, except it is the document currently being edited that is printed, rather than a document stored in the database.
You can exit the application by selecting 'File->Exit' from the application menu. If you are not using 'Always-online' mode and you still have local documents that need to be committed, Collect will prompt you to do so with the following dialog.
It is recommended that you commit your data to the database frequently.